Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Low-complexity nonlinear adaptive filter based on a pipelined bilinear recurrent neural network.

Haiquan Zhao1, Xiangping Zeng, Zhengyou He

  • 1School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China. hqzhao@home.swjtu.edu.cn

IEEE Transactions on Neural Networks
|August 2, 2011
PubMed
Summary

A new Pipelined Bilinear Recurrent Neural Network (PBLRNN) significantly reduces computational complexity for adaptive filters. This novel approach enhances efficiency and performance in nonlinear system identification and prediction tasks.

Related Concept Videos

Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Optimization of Ultrasound-Assisted Deep Eutectic Solvent Extraction and Mechanism Evaluation of Saponins from <i>Panax japonicus</i>.

Molecules (Basel, Switzerland)·2026
Same author

Enhanced anti-hyperuricemic effect of Simiao pill by composite lactic acid bacteria fermentation: ferulic acid and berberine enrichment and URAT1/OAT1/XOD regulation.

Frontiers in pharmacology·2026
Same author

Qing'e Pill extract attenuates vascular calcification and improves renal dysfunction in chronic kidney disease mice by modulating calcium-phosphorus homeostasis and downregulating the expression of Runx2/OCN.

Journal of ethnopharmacology·2026
Same author

Causal Effects of a Hepatic Senescence Gene Set on MASLD Fibrosis: A Mendelian Randomization Study and Quercetin Molecular Docking Analysis.

Biomedicines·2026
Same author

Amorphous IrO<sub>x</sub> buffer layer stabilizes low-valence Ir centers and reinforces IrO hybridization in hierarchical multilayered Ir-IrCoO<sub>x</sub> for acidic oxygen evolution.

Journal of colloid and interface science·2026
Same author

Engineering Crystalline-Amorphous Interfaces in Iron Sulfide/Carbon Nanostructures to Boost Sodium Storage Kinetics and Cycling Durability.

ACS applied materials & interfaces·2025

Area of Science:

  • Signal Processing
  • Machine Learning
  • Artificial Neural Networks

Background:

  • Bilinear Recurrent Neural Networks (BLRNNs) offer powerful nonlinear modeling capabilities but suffer from high computational complexity.
  • Existing recurrent neural network (RNN) architectures often struggle with efficiency in complex nonlinear adaptive filtering tasks.

Purpose of the Study:

  • To introduce a novel low-complexity nonlinear adaptive filter using a Pipelined Bilinear Recurrent Neural Network (PBLRNN).
  • To enhance computational efficiency and performance in nonlinear adaptive filtering applications.

Main Methods:

  • The PBLRNN is designed with modular, chained BLRNN units, enabling pipelined parallelism for improved computational efficiency.
  • A modified adaptive amplitude real-time recurrent learning algorithm based on gradient descent is developed for the modular architecture.

Related Experiment Videos

  • The model's effectiveness is evaluated through extensive simulations on nonlinear system identification, channel equalization, and chaotic time series prediction.

    • Main Results:

    • The PBLRNN demonstrates significantly improved computational efficiency compared to standard BLRNN and RNN models.
    • Experimental results confirm superior performance of the PBLRNN across various nonlinear adaptive filtering tasks.
    • Nesting modules within the PBLRNN architecture further enhances its predictive and adaptive capabilities.

    Conclusions:

    • The Pipelined Bilinear Recurrent Neural Network (PBLRNN) offers a computationally efficient and high-performing solution for nonlinear adaptive filtering.
    • The modular and pipelined design of PBLRNN overcomes the complexity limitations of traditional BLRNNs.
    • PBLRNN shows considerable promise for applications in nonlinear system identification, channel equalization, and chaotic time series prediction.